Optimization of Playful Learning on the UnispherTM Platform by Simulation Modeling

Abstract

Education optimization of playful learning student`s groups is considered. The goal of optimization is to form groups of students who will show high performance of education using game forms of learning. The optimization is based on the student`s classification according to interest and learning ability, behav-ioral characteristics, character traits and social characteristics, as well as on predicting their individual and group performance, considering the mutual influence and work of teachers and mentors. The stu-dents are presented by vectors of their feature`s numerical assessments. The student`s classification is carried out using the method of Independent Component Analysis on a data driven basis obtained by Singular Value Decomposition. The method allows to classify students` data taking into account the sig-nificance of the parameters and an uncertainty of some parameter`s values through their randomization. The method allows to students with different learning abilities and behavioral characteristics for each group in order to create conditions for the coordinated work of the group. The method allows to dis-tribute students with special characteristics evenly into groups.